1. Field of the Invention
The invention disclosed is broadly related to telecommunications and is more particularly related to systems and methods to provide survivability for cable access networks.
2. Prior Art
Broadband access networks (e.g. hybrid fiber/coaxial) are used to provide television programming and Internet access for customers. More recently they are being investigated to provide IP telephony service. However, the reliability of hybrid fiber/coaxial (HFC) cable networks is not comparable to the high reliability of the local telephone network. Local telephone networks are designed to meet an unavailability objective of 53 minutes/year/line. HFC cable networks have a downtime of 622 minutes/year/line due to AC power failure and additional 125 minutes/year/line due to failures other than AC power failure (e.g. amplifier failure, water leakage, cable cuts). One key difference between a local loop network and an HFC cable network is that the former uses battery power at the central office. The latter requires commercial AC power in the field for amplifiers and coaxial taps. Loss of AC power causes a failure of the HFC cable delivery network.
What is needed is a cost-effective approach to significantly improve the overall downtime of an HFC cable network from its current (xcx9c747/min/yr/line). What is needed is a way to implement physically diverse-path routing (self-healing) for the information-bearing signals (e.g., video, Internet access, IP telephony) to improve the xe2x80x9csurvivabilityxe2x80x9d of the HFC network. What is needed is a way to improve downtime by providing protection against single point failures of one or more elements constituting the fiber/coax link, including the cables, amplifiers, taps, etc.
A communications network topology is disclosed that provides survivability for broadcast video and audio and interactive data communications services on cable access networks. A hybrid fiber/coaxial cable connected to a headend or distribution hub, provides connectivity for many cable data modems and set top boxes in a neighborhood or community, in the conventional manner. When a cable""s connection to the headend is accidentally interrupted, service can be rapidly restored to the affected cable data modems and set top boxes attached to the cable, by means of the invention disclosed herein.
In accordance with the invention, the hybrid fiber/coaxial cable is attached to an autonomous controller which includes a wireless receiver, transmitter, and antenna. Typically, the autonomous controller is connected near the end of the cable which is remote from the headend. At least one other hybrid fiber/coaxial cable connected to the same headend or distribution hub, provides connectivity for cable data modems and set top boxes in an adjacent neighborhood or community. The second cable is attached to its own, respective autonomous controller which also has a wireless receiver, transmitter, and antenna. The two antennas in the adjacent neighborhoods constitute an antenna cluster. When the first cable""s connection to the common headend is interrupted, thereby preventing service to the first neighborhood, its autonomous controller senses the interruption and responds by sending a wireless signal to the controller in the adjacent neighborhood. In response, the two controllers use an antenna cluster protocol (ACP) to set up a wireless communications path through the second cable to the headend. In this manner, service is restored from the headend to the cable data modems and set top boxes in the first neighborhood.
The antenna cluster protocol (ACP) is based on each autonomous controller including a state machine having a standby state, a protected state, and an active state. During the normal operation of the cable, the controller detects signals arriving from the headend via the cable, and the state machine in the controller uses this indication to remain in the standby state. When in the standby state, the wireless receiver in the controller continues operating to receive any signals on an upstream frequency from other antennas in its cluster, but the wireless transmitter is turned off to conserve power.
When the signals from the headend through the first cable fail to be received at the first controller, the first state machine changes to the protected state. In the protected state, the controller""s transmitter is turned on and an ACP control packet is transmitted to other antennas in its cluster on the upstream frequency. This ACP control packet identifies the sender as the first controller and it indicates that the first controller is in the protected state. Since any other controllers in the same cluster will be receiving on the same frequency, they are immediately made aware of the outage condition that has occurred on the first cable. Upon changing to the protected state, the wireless receiver in the first controller changes its receiving frequency to a downstream frequency.
Each controller in an antenna cluster is preassigned a distinct time interval value to wait while in the standby state before responding to the receipt of the ACP protected state control packet. The state machine in the controller having the shortest waiting interval changes from the standby to the active state. Thus, in response to having received the ACP protected state control packet, the second controller changes to the active state. In the active state, the second controller""s transmitter is turned on and an ACP control packet is transmitted to other antennas in its cluster on the downstream frequency. This ACP control packet identifies the sender as the second controller and it indicates that the second controller is in the active state. Since only controllers that are in the protected state can receive on the downstream frequency, the protected state controllers are immediately made aware that there is a controller available to provide an alternate path to the headend to circumvent the outage condition that has occurred on the first cable.
Once the first and second controllers identify themselves to each other as being in the protected and active states, respectively, the antenna cluster protocol (ACP) enables traffic signals to pass to and from the first controller via its wireless antenna to the wireless antenna of the second controller for transmission through the second cable path to and from the common headend. In this manner, the outage condition that has occurred on the first cable is circumvented.
In antenna clusters consisting of more than two controllers connected through their respective cables to a common headend, the active state controller can exchange the traffic signals with the headend for a plurality of protected state controllers through the active controller""s cable path.
To prevent more than one controller at a time from remaining in the active state, each of the controllers has a unique identity and priority for remaining in the active state. If an active state controller determines that there is another active controller having a higher priority, then the lower priority controller reverts to the standby state.
A controller in the protected state periodically broadcasts an ACP protected state control packet until its cable is repaired or it is taken off-line. The active state controller monitors these ACP control signals, and when they no longer are received, the active state controller reverts to the standby state.
A controller in the protected state continues to monitor for the resumption of the signals arriving from the headend via its cable, and its state machine uses this condition of an absence of such signals to remain in the protected state. When the signals from the headend through the cable are once again received at the protected state controller, it changes to the standby state.
An active state controller continues to detect the signals arriving from the headend via its cable, and its state machine uses this indication to remain in the active state. When the signals from the headend through the cable fail to be received at the active state controller, it changes to the protected state. Then another, standby state controller in the same cluster will change its state to the active state. The antenna cluster protocol (ACP) enables traffic signals to and from the old active state (now protected state) controller to pass via their wireless antennas to the new active state controller for transmission through its cable path to and from the common headend.
In one preferred embodiment, the ACP protected state control packet is an ICMP message broadcast via the wireless antenna of the controller in the protected state and the ACP active state control packet is an ICMP message broadcast via the wireless antenna of the controller in the active state.
In one preferred embodiment, the signals from the cable data modems are encrypted before being transmitted by the wireless antenna of the protected state controller and the signals from the headend are encrypted before being transmitted by the wireless antenna of active state controller.
In one the preferred embodiment, the ACP protected state control packet and the ACP active state control packet are not sent to the headend.
In another preferred embodiment, a protected state ICMP alarm message is sent to the headend to provide notice that the cable path of the protected controller has failed.
In still another preferred embodiment, a protected state SNMP trap PDU alert message is sent to a network monitor to provide notice that the cable path of the protected controller has failed.